Papers with Code · CC BY-SA 4.0

Civil Comments

아카이브 평가 22건 · 마지막 적재/변경 2026-10-30

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
RoBERTa Focal Loss	AUROC 0.9818 Macro F1 0.4648 Micro F1 0.5524 Precision 0.4017 Recall 0.8839 GMB Subgroup 0.8807 GMB BPSN 0.901 GMB BNSP 0.9581	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
AlBERT	AUROC 0.979 Macro F1 0.3541 Micro F1 0.4845 Precision 0.3247 Recall 0.9104 GMB Subgroup 0.8734 GMB BPSN 0.8982 GMB BNSP 0.9499	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
BERTweet	AUROC 0.979 Macro F1 0.3612 Micro F1 0.4928 Precision 0.3363 Recall 0.9216 GMB Subgroup 0.878 GMB BPSN 0.8945 GMB BNSP 0.9603	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
HateBERT	AUROC 0.9791 Macro F1 0.3679 Micro F1 0.4844 Precision 0.3297 Recall 0.9165 GMB Subgroup 0.8744 GMB BPSN 0.8915 GMB BNSP 0.9589	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
RoBERTa BCE	AUROC 0.9813 Macro F1 0.4749 Micro F1 0.5359 Precision 0.3836 Recall 0.8891 GMB Subgroup 0.88 GMB BPSN 0.8901 GMB BNSP 0.9616	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
XLM RoBERTa	Micro F1 0.468 Precision 0.3135 Recall 0.923 GMB BPSN 0.8859	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
XLNet	Macro F1 0.3336 Micro F1 0.4586 Precision 0.3045 Recall 0.9254 GMB Subgroup 0.8689 GMB BPSN 0.8834 GMB BNSP 0.9597	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
DistilBERT	AUROC 0.9804 Macro F1 0.3879 Micro F1 0.5115 Precision 0.3572 Recall 0.9001 GMB Subgroup 0.8762 GMB BPSN 0.874 GMB BNSP 0.9644	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
BiGRU	GMB BPSN 0.8616	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
Unfreeze Glove ResNet 44	AUROC 0.966 Macro F1 0.4648 Micro F1 0.5958 Precision 0.4835 Recall 0.7759 GMB Subgroup 0.8421 GMB BPSN 0.8493	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
Unfreeze Glove ResNet 56	AUROC 0.9639 Macro F1 0.3778 Recall 0.8707 GMB Subgroup 0.8487 GMB BPSN 0.8445	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
Compact Convolutional Transformer (CCT)	AUROC 0.9526 Macro F1 0.3428 Micro F1 0.4874 Precision 0.3507 Recall 0.7983 GMB Subgroup 0.8133 GMB BPSN 0.8307 GMB BNSP 0.9447	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
Freeze Glove ResNet 44	Macro F1 0.4189 Micro F1 0.5591 Precision 0.4631 Recall 0.7053 GMB Subgroup 0.8219 GMB BPSN 0.7876	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
BiLSTM	Micro F1 0.5115 Precision 0.3572 GMB Subgroup 0.8636	A benchmark for toxic comment classification on Civil Comments dataset 2023-01-26	코드 1
ResNet + RoBERTa finetune	AUROC 0.97	PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning 2024-03-31	코드 1
Trompt + OpenAI embedding	AUROC 0.947	PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning 2024-03-31	코드 1
ResNet + OpenAI embedding	AUROC 0.945	PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning 2024-03-31	코드 1
Trompt + RoBERTa embedding	AUROC 0.885	PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning 2024-03-31	코드 1
ResNet + RoBERTa embedding	AUROC 0.882	PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning 2024-03-31	코드 1
LightGBM + RoBERTa embedding	AUROC 0.865	PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning 2024-03-31	코드 1
PaLM 2 (few-shot, k=10)	AUROC 0.8535	PaLM 2 Technical Report 2023-05-17	코드 1
PaLM 2 (zero-shot)	AUROC 0.7596	PaLM 2 Technical Report 2023-05-17	코드 1